{
  "anchors": {
    "agent_arena_ips_score": {
      "a": -0.25,
      "b": 0.25,
      "direction": "higher"
    }
  },
  "as_of": "2026-11-05",
  "attributions": [
    {
      "source_id": "arena-leaderboard-dataset",
      "text": "Arena data: Arena (lmarena-ai), dataset lmarena-ai/leaderboard-dataset, https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset, licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Changes by Inadviser: the Arena scores (the Agent Arena IPS score and the text, text-to-image, text-to-video and search ratings) are rescaled to a 0-100 utility against fixed anchors set by Inadviser; the confidence interval and the observation or vote count are carried over; only a selection of models is shown on each board; where a board shows Arena's cost per task, it is shown unchanged. The dataset maintainers' method applies (https://arena.ai/blog/agent-arena-methodology/ for Agent Arena). Arena does not endorse Inadviser."
    }
  ],
  "board": {
    "category": "ai-agents",
    "cost": {
      "kind": "metric",
      "label": "Median cost per task",
      "metric": "cost_per_task_p50",
      "unit": "USD per task (median)"
    },
    "half_life_days": 30,
    "job": "Run an AI agent on real software tasks and judge it by how much the agent's own work was kept and corrected, against what each task costs.",
    "price_limit_days": 30,
    "products": [
      {
        "id": "claude-opus-5-5-max",
        "name": "Claude Opus 5.5 (Max)",
        "vendor": "anthropic"
      },
      {
        "id": "gemini-4-carbon-high",
        "name": "Gemini 4 Carbon (High)",
        "vendor": "google"
      },
      {
        "id": "claude-sonnet-5-5-max",
        "name": "Claude Sonnet 5.5 (Max)",
        "vendor": "anthropic"
      },
      {
        "id": "gpt-6-astra-max",
        "name": "GPT-6 Astra (Max)",
        "vendor": "openai"
      },
      {
        "id": "claude-fable-5-1-max",
        "name": "Claude Fable 5.1 (Max)",
        "vendor": "anthropic"
      },
      {
        "id": "gpt-6-1-sol-max",
        "name": "GPT-6.1 Sol (Max)",
        "vendor": "openai"
      },
      {
        "id": "gemini-4-argon-high",
        "name": "Gemini 4 Argon (High)",
        "vendor": "google"
      },
      {
        "id": "gpt-6-1-terra-max",
        "name": "GPT-6.1 Terra (Max)",
        "vendor": "openai"
      },
      {
        "id": "muse-wonder-full-max",
        "name": "Muse Wonder Full (Max)",
        "vendor": "meta"
      },
      {
        "id": "claude-fable-5-high",
        "name": "Claude Fable 5 (High)",
        "vendor": "anthropic"
      },
      {
        "id": "claude-opus-5-max",
        "name": "Claude Opus 5 (Max)",
        "vendor": "anthropic"
      },
      {
        "id": "gpt-6-sol-max",
        "name": "GPT-6 Sol (Max)",
        "vendor": "openai"
      },
      {
        "id": "kimi-k3-1-pro-max",
        "name": "Kimi K3.1 Pro (Max)",
        "vendor": "moonshot"
      },
      {
        "id": "gpt-5-6-sol-xhigh",
        "name": "GPT-5.6 Sol (xHigh)",
        "vendor": "openai"
      },
      {
        "id": "claude-sonnet-5-high",
        "name": "Claude Sonnet 5 (High)",
        "vendor": "anthropic"
      },
      {
        "id": "grok-4-7-xhigh",
        "name": "Grok 4.7 (xHigh)",
        "vendor": "xai"
      },
      {
        "id": "gemini-3-8-flash-high",
        "name": "Gemini 3.8 Flash (High)",
        "vendor": "google"
      },
      {
        "id": "muse-wonder-max",
        "name": "Muse Wonder (Max)",
        "vendor": "meta"
      },
      {
        "id": "muse-spark-1-3-max",
        "name": "Muse Spark 1.3 (Max)",
        "vendor": "meta"
      },
      {
        "id": "mimo-v2-6-pro",
        "name": "MiMo-V2.6-Pro",
        "vendor": "xiaomi"
      },
      {
        "id": "deepseek-v4-1-flash-20261017-max",
        "name": "DeepSeek-V4.1-Flash-20261017 (Max)",
        "vendor": "deepseek"
      },
      {
        "id": "gpt-6-luna-max",
        "name": "GPT-6 Luna (Max)",
        "vendor": "openai"
      },
      {
        "id": "glm-5-3-max",
        "name": "GLM-5.3 (Max)",
        "vendor": "zai"
      },
      {
        "id": "hy4-preview",
        "name": "Hy4 preview",
        "vendor": "tencent"
      },
      {
        "id": "kimi-k3-max",
        "name": "Kimi K3 (Max)",
        "vendor": "moonshot"
      },
      {
        "id": "glm-5-2-max",
        "name": "GLM-5.2 (Max)",
        "vendor": "zai"
      },
      {
        "id": "claude-haiku-5-5-high",
        "name": "Claude Haiku 5.5 (High)",
        "vendor": "anthropic"
      },
      {
        "id": "mimo-v2-6-flash",
        "name": "MiMo-V2.6-Flash",
        "vendor": "xiaomi"
      },
      {
        "id": "qwen3-8-max",
        "name": "Qwen3.8-Max",
        "vendor": "alibaba"
      },
      {
        "id": "grok-4-6-xhigh",
        "name": "Grok 4.6 (xHigh)",
        "vendor": "xai"
      },
      {
        "id": "deepseek-v4-pro-high-20260813",
        "name": "DeepSeek-V4-Pro (High, 20260813)",
        "vendor": "deepseek"
      },
      {
        "id": "gpt-5-6-terra-xhigh",
        "name": "GPT-5.6 Terra (xHigh)",
        "vendor": "openai"
      },
      {
        "id": "gemini-3-7-flash-high",
        "name": "Gemini 3.7 Flash (High)",
        "vendor": "google"
      },
      {
        "id": "gpt-6-luna-medium",
        "name": "GPT-6 Luna (Medium)",
        "vendor": "openai"
      },
      {
        "id": "glm-5-3-flash",
        "name": "GLM-5.3-Flash",
        "vendor": "zai"
      },
      {
        "id": "step-5-preview",
        "name": "Step 5 Preview",
        "vendor": "stepfun"
      },
      {
        "id": "minimax-m3-1-flash",
        "name": "MiniMax-M3.1-Flash",
        "vendor": "minimax"
      },
      {
        "id": "qwen3-8-27b",
        "name": "Qwen3.8-27B",
        "vendor": "alibaba"
      },
      {
        "id": "qwen3-8-flash-next",
        "name": "Qwen3.8-Flash-Next",
        "vendor": "alibaba"
      },
      {
        "id": "gemini-3-1-pro-preview",
        "name": "Gemini 3.1 Pro Preview",
        "vendor": "google"
      },
      {
        "id": "beam-1-0-501b-a23b",
        "name": "Beam 1.0 501B A23B",
        "vendor": "reflection"
      },
      {
        "id": "gpt-6-luna-low",
        "name": "GPT-6 Luna (Low)",
        "vendor": "openai"
      },
      {
        "id": "mistral-large-4",
        "name": "Mistral Large 4",
        "vendor": "mistral"
      },
      {
        "id": "qwen3-7-max",
        "name": "Qwen3.7-Max",
        "vendor": "alibaba"
      },
      {
        "id": "qwen3-7-plus",
        "name": "Qwen3.7-Plus",
        "vendor": "alibaba"
      },
      {
        "id": "hy3",
        "name": "Hy3",
        "vendor": "tencent"
      },
      {
        "id": "muse-spark-1-1",
        "name": "Muse Spark 1.1",
        "vendor": "meta"
      },
      {
        "id": "muse-spark-1-2-xhigh",
        "name": "Muse Spark 1.2 (xHigh)",
        "vendor": "meta"
      },
      {
        "id": "mimo-v2-5-pro",
        "name": "MiMo-V2.5-Pro",
        "vendor": "xiaomi"
      },
      {
        "id": "spark-x2-6",
        "name": "Spark-X2.6",
        "vendor": "synlan"
      },
      {
        "id": "minimax-m3",
        "name": "MiniMax-M3",
        "vendor": "minimax"
      },
      {
        "id": "mistral-medium-3-5",
        "name": "Mistral Medium 3.5",
        "vendor": "mistral"
      },
      {
        "id": "nemotron-3-ultra",
        "name": "Nemotron 3 Ultra",
        "vendor": "nvidia"
      },
      {
        "id": "nvidia-nemotron-3-5-super-vl-120b-a12b-nvfp4",
        "name": "NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)",
        "vendor": "nvidia"
      },
      {
        "id": "inkling",
        "name": "Inkling",
        "vendor": "thinky"
      },
      {
        "id": "inkling-small",
        "name": "Inkling-Small",
        "vendor": "thinky"
      },
      {
        "id": "solar-pro-4",
        "name": "Solar Pro 4",
        "vendor": "upstage"
      }
    ],
    "question": "Which AI model gives the most useful agent work for the least cost per task?",
    "region": "global",
    "requirements": {},
    "slug": "ai-agents",
    "title": "AI coding agents",
    "workload": {
      "chars_per_month": null,
      "unit_note": "One task. Cost is the median cost per task that Arena reports for the model."
    }
  },
  "code_commit": "6585be527486036d41ec57c349ffe2a97aa8cd24",
  "decision": {
    "disqualified": [],
    "leader": "claude-opus-5-5-max",
    "margin": 3,
    "needed": [
      {
        "metric": "agent_arena_ips_score",
        "need": "another independent origin or repeats",
        "product": "claude-opus-5-5-max",
        "type": "narrow_interval",
        "width": 8.447762
      },
      {
        "metric": "agent_arena_ips_score",
        "need": "another independent origin or repeats",
        "product": "claude-sonnet-5-5-max",
        "type": "narrow_interval",
        "width": 11.140867
      },
      {
        "metric": "agent_arena_ips_score",
        "need": "another independent origin or repeats",
        "product": "gemini-4-carbon-high",
        "type": "narrow_interval",
        "width": 3.222774
      }
    ],
    "reasons": [
      {
        "code": "bounds_overlap",
        "gap": 1.045983,
        "other": "claude-sonnet-5-5-max",
        "product": "claude-opus-5-5-max",
        "required": 3
      },
      {
        "code": "bounds_overlap",
        "gap": 1.123185,
        "other": "gemini-4-carbon-high",
        "product": "claude-opus-5-5-max",
        "required": 3
      }
    ],
    "shortlist": [
      "claude-opus-5-5-max",
      "gemini-4-carbon-high",
      "claude-sonnet-5-5-max"
    ],
    "unmeasured": [],
    "verdict": "shortlist",
    "winner": null
  },
  "evidence_ids": [
    "arena-agent.beam-1-0-501b-a23b.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-fable-5-1-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-fable-5-1-max.cost_per_task_p50.2026-11-02",
    "arena-agent.claude-fable-5-high.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-fable-5-high.cost_per_task_p50.2026-11-02",
    "arena-agent.claude-haiku-5-5-high.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-haiku-5-5-high.cost_per_task_p50.2026-11-02",
    "arena-agent.claude-opus-5-5-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-opus-5-5-max.cost_per_task_p50.2026-11-02",
    "arena-agent.claude-opus-5-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-opus-5-max.cost_per_task_p50.2026-11-02",
    "arena-agent.claude-sonnet-5-5-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-sonnet-5-5-max.cost_per_task_p50.2026-11-02",
    "arena-agent.claude-sonnet-5-high.agent_arena_ips_score.2026-11-02",
    "arena-agent.claude-sonnet-5-high.cost_per_task_p50.2026-11-02",
    "arena-agent.deepseek-v4-1-flash-20261017-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.deepseek-v4-1-flash-20261017-max.cost_per_task_p50.2026-11-02",
    "arena-agent.deepseek-v4-pro-high-20260813.agent_arena_ips_score.2026-11-02",
    "arena-agent.gemini-3-1-pro-preview.agent_arena_ips_score.2026-11-02",
    "arena-agent.gemini-3-1-pro-preview.cost_per_task_p50.2026-11-02",
    "arena-agent.gemini-3-7-flash-high.agent_arena_ips_score.2026-11-02",
    "arena-agent.gemini-3-7-flash-high.cost_per_task_p50.2026-11-02",
    "arena-agent.gemini-3-8-flash-high.agent_arena_ips_score.2026-11-02",
    "arena-agent.gemini-3-8-flash-high.cost_per_task_p50.2026-11-02",
    "arena-agent.gemini-4-argon-high.agent_arena_ips_score.2026-11-02",
    "arena-agent.gemini-4-argon-high.cost_per_task_p50.2026-11-02",
    "arena-agent.gemini-4-carbon-high.agent_arena_ips_score.2026-11-02",
    "arena-agent.gemini-4-carbon-high.cost_per_task_p50.2026-11-02",
    "arena-agent.glm-5-2-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.glm-5-2-max.cost_per_task_p50.2026-11-02",
    "arena-agent.glm-5-3-flash.agent_arena_ips_score.2026-11-02",
    "arena-agent.glm-5-3-flash.cost_per_task_p50.2026-11-02",
    "arena-agent.glm-5-3-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.glm-5-3-max.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-5-6-sol-xhigh.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-5-6-sol-xhigh.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-5-6-terra-xhigh.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-5-6-terra-xhigh.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-6-1-sol-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-6-1-sol-max.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-6-1-terra-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-6-1-terra-max.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-6-astra-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-6-astra-max.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-6-luna-low.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-6-luna-low.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-6-luna-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-6-luna-max.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-6-luna-medium.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-6-luna-medium.cost_per_task_p50.2026-11-02",
    "arena-agent.gpt-6-sol-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.gpt-6-sol-max.cost_per_task_p50.2026-11-02",
    "arena-agent.grok-4-6-xhigh.agent_arena_ips_score.2026-11-02",
    "arena-agent.grok-4-6-xhigh.cost_per_task_p50.2026-11-02",
    "arena-agent.grok-4-7-xhigh.agent_arena_ips_score.2026-11-02",
    "arena-agent.grok-4-7-xhigh.cost_per_task_p50.2026-11-02",
    "arena-agent.hy3.agent_arena_ips_score.2026-11-02",
    "arena-agent.hy3.cost_per_task_p50.2026-11-02",
    "arena-agent.hy4-preview.agent_arena_ips_score.2026-11-02",
    "arena-agent.hy4-preview.cost_per_task_p50.2026-11-02",
    "arena-agent.inkling-small.agent_arena_ips_score.2026-11-02",
    "arena-agent.inkling-small.cost_per_task_p50.2026-11-02",
    "arena-agent.inkling.agent_arena_ips_score.2026-11-02",
    "arena-agent.inkling.cost_per_task_p50.2026-11-02",
    "arena-agent.kimi-k3-1-pro-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.kimi-k3-1-pro-max.cost_per_task_p50.2026-11-02",
    "arena-agent.kimi-k3-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.kimi-k3-max.cost_per_task_p50.2026-11-02",
    "arena-agent.mimo-v2-5-pro.agent_arena_ips_score.2026-11-02",
    "arena-agent.mimo-v2-5-pro.cost_per_task_p50.2026-11-02",
    "arena-agent.mimo-v2-6-flash.agent_arena_ips_score.2026-11-02",
    "arena-agent.mimo-v2-6-flash.cost_per_task_p50.2026-11-02",
    "arena-agent.mimo-v2-6-pro.agent_arena_ips_score.2026-11-02",
    "arena-agent.mimo-v2-6-pro.cost_per_task_p50.2026-11-02",
    "arena-agent.minimax-m3-1-flash.agent_arena_ips_score.2026-11-02",
    "arena-agent.minimax-m3-1-flash.cost_per_task_p50.2026-11-02",
    "arena-agent.minimax-m3.agent_arena_ips_score.2026-11-02",
    "arena-agent.minimax-m3.cost_per_task_p50.2026-11-02",
    "arena-agent.mistral-large-4.agent_arena_ips_score.2026-11-02",
    "arena-agent.mistral-large-4.cost_per_task_p50.2026-11-02",
    "arena-agent.mistral-medium-3-5.agent_arena_ips_score.2026-11-02",
    "arena-agent.mistral-medium-3-5.cost_per_task_p50.2026-11-02",
    "arena-agent.muse-spark-1-1.agent_arena_ips_score.2026-11-02",
    "arena-agent.muse-spark-1-1.cost_per_task_p50.2026-11-02",
    "arena-agent.muse-spark-1-2-xhigh.agent_arena_ips_score.2026-11-02",
    "arena-agent.muse-spark-1-2-xhigh.cost_per_task_p50.2026-11-02",
    "arena-agent.muse-spark-1-3-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.muse-spark-1-3-max.cost_per_task_p50.2026-11-02",
    "arena-agent.muse-wonder-full-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.muse-wonder-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.muse-wonder-max.cost_per_task_p50.2026-11-02",
    "arena-agent.nemotron-3-ultra.agent_arena_ips_score.2026-11-02",
    "arena-agent.nemotron-3-ultra.cost_per_task_p50.2026-11-02",
    "arena-agent.nvidia-nemotron-3-5-super-vl-120b-a12b-nvfp4.agent_arena_ips_score.2026-11-02",
    "arena-agent.qwen3-7-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.qwen3-7-max.cost_per_task_p50.2026-11-02",
    "arena-agent.qwen3-7-plus.agent_arena_ips_score.2026-11-02",
    "arena-agent.qwen3-7-plus.cost_per_task_p50.2026-11-02",
    "arena-agent.qwen3-8-27b.agent_arena_ips_score.2026-11-02",
    "arena-agent.qwen3-8-27b.cost_per_task_p50.2026-11-02",
    "arena-agent.qwen3-8-flash-next.agent_arena_ips_score.2026-11-02",
    "arena-agent.qwen3-8-flash-next.cost_per_task_p50.2026-11-02",
    "arena-agent.qwen3-8-max.agent_arena_ips_score.2026-11-02",
    "arena-agent.qwen3-8-max.cost_per_task_p50.2026-11-02",
    "arena-agent.solar-pro-4.agent_arena_ips_score.2026-11-02",
    "arena-agent.solar-pro-4.cost_per_task_p50.2026-11-02",
    "arena-agent.spark-x2-6.agent_arena_ips_score.2026-11-02",
    "arena-agent.spark-x2-6.cost_per_task_p50.2026-11-02",
    "arena-agent.step-5-preview.agent_arena_ips_score.2026-11-02",
    "arena-agent.step-5-preview.cost_per_task_p50.2026-11-02"
  ],
  "pareto": {
    "dominated": [
      {
        "dominated_by": [
          "claude-opus-5-5-max",
          "claude-sonnet-5-5-max",
          "gemini-4-carbon-high",
          "gpt-6-astra-max"
        ],
        "id": "claude-fable-5-1-max"
      },
      {
        "dominated_by": [
          "gemini-4-argon-high",
          "gemini-4-carbon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max"
        ],
        "id": "claude-fable-5-high"
      },
      {
        "dominated_by": [
          "gpt-6-luna-max"
        ],
        "id": "claude-haiku-5-5-high"
      },
      {
        "dominated_by": [
          "claude-fable-5-high",
          "claude-sonnet-5-5-max",
          "gemini-4-argon-high",
          "gemini-4-carbon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max"
        ],
        "id": "claude-opus-5-max"
      },
      {
        "dominated_by": [
          "gemini-4-carbon-high"
        ],
        "id": "claude-sonnet-5-5-max"
      },
      {
        "dominated_by": [
          "gemini-4-argon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max"
        ],
        "id": "claude-sonnet-5-high"
      },
      {
        "dominated_by": [
          "mimo-v2-6-pro"
        ],
        "id": "deepseek-v4-1-flash-20261017-max"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gemini-3-7-flash-high",
          "glm-5-3-flash",
          "gpt-5-6-terra-xhigh",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy4-preview",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3-1-flash",
          "muse-spark-1-3-max",
          "qwen3-8-27b",
          "qwen3-8-flash-next",
          "step-5-preview"
        ],
        "id": "gemini-3-1-pro-preview"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gpt-6-luna-max",
          "hy4-preview",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max"
        ],
        "id": "gemini-3-7-flash-high"
      },
      {
        "dominated_by": [
          "gpt-6-1-sol-max"
        ],
        "id": "gemini-4-argon-high"
      },
      {
        "dominated_by": [
          "deepseek-v4-1-flash-20261017-max",
          "gemini-3-8-flash-high",
          "gemini-4-argon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max",
          "gpt-6-luna-max",
          "hy4-preview",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max"
        ],
        "id": "glm-5-2-max"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro"
        ],
        "id": "glm-5-3-flash"
      },
      {
        "dominated_by": [
          "claude-sonnet-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gemini-3-8-flash-high",
          "gemini-4-argon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max",
          "gpt-6-luna-max",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max"
        ],
        "id": "glm-5-3-max"
      },
      {
        "dominated_by": [
          "gemini-4-argon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max",
          "kimi-k3-1-pro-max"
        ],
        "id": "gpt-5-6-sol-xhigh"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gpt-6-luna-max",
          "hy4-preview",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max"
        ],
        "id": "gpt-5-6-terra-xhigh"
      },
      {
        "dominated_by": [
          "claude-sonnet-5-5-max",
          "gemini-4-carbon-high"
        ],
        "id": "gpt-6-astra-max"
      },
      {
        "dominated_by": [
          "gemini-4-argon-high",
          "gemini-4-carbon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max"
        ],
        "id": "gpt-6-sol-max"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "claude-sonnet-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gemini-3-8-flash-high",
          "gemini-4-argon-high",
          "glm-5-2-max",
          "glm-5-3-max",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max",
          "gpt-6-luna-max",
          "hy4-preview",
          "kimi-k3-1-pro-max",
          "kimi-k3-max",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max",
          "muse-wonder-max",
          "qwen3-8-max"
        ],
        "id": "grok-4-6-xhigh"
      },
      {
        "dominated_by": [
          "claude-sonnet-5-high",
          "gemini-4-argon-high",
          "gemini-4-carbon-high",
          "gpt-5-6-sol-xhigh",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max",
          "gpt-6-sol-max",
          "kimi-k3-1-pro-max"
        ],
        "id": "grok-4-7-xhigh"
      },
      {
        "dominated_by": [
          "gpt-6-luna-low",
          "gpt-6-luna-medium",
          "mimo-v2-6-flash"
        ],
        "id": "hy3"
      },
      {
        "dominated_by": [
          "deepseek-v4-1-flash-20261017-max",
          "gpt-6-luna-max",
          "mimo-v2-6-pro"
        ],
        "id": "hy4-preview"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-5-pro",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3",
          "minimax-m3-1-flash",
          "nemotron-3-ultra",
          "qwen3-7-plus",
          "qwen3-8-flash-next",
          "spark-x2-6"
        ],
        "id": "inkling"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-5-pro",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "nemotron-3-ultra",
          "qwen3-7-plus",
          "spark-x2-6"
        ],
        "id": "inkling-small"
      },
      {
        "dominated_by": [
          "gemini-4-argon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max"
        ],
        "id": "kimi-k3-1-pro-max"
      },
      {
        "dominated_by": [
          "claude-sonnet-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gemini-3-8-flash-high",
          "gemini-4-argon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max",
          "gpt-6-luna-max",
          "hy4-preview",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max"
        ],
        "id": "kimi-k3-max"
      },
      {
        "dominated_by": [
          "gpt-6-luna-low",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-6-flash"
        ],
        "id": "mimo-v2-5-pro"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-5-pro",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "qwen3-7-plus",
          "qwen3-8-flash-next",
          "spark-x2-6"
        ],
        "id": "minimax-m3"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro"
        ],
        "id": "minimax-m3-1-flash"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy4-preview",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3-1-flash",
          "qwen3-8-27b",
          "qwen3-8-flash-next"
        ],
        "id": "mistral-large-4"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "hy4-preview",
          "mimo-v2-5-pro",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3",
          "minimax-m3-1-flash",
          "mistral-large-4",
          "muse-spark-1-1",
          "muse-spark-1-2-xhigh",
          "muse-spark-1-3-max",
          "qwen3-7-max",
          "qwen3-7-plus",
          "qwen3-8-27b",
          "qwen3-8-flash-next",
          "spark-x2-6"
        ],
        "id": "mistral-medium-3-5"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3-1-flash",
          "qwen3-7-max",
          "qwen3-7-plus",
          "qwen3-8-flash-next"
        ],
        "id": "muse-spark-1-1"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3-1-flash",
          "muse-spark-1-1",
          "qwen3-7-max",
          "qwen3-7-plus",
          "qwen3-8-27b",
          "qwen3-8-flash-next"
        ],
        "id": "muse-spark-1-2-xhigh"
      },
      {
        "dominated_by": [
          "claude-sonnet-5-high",
          "gemini-3-8-flash-high",
          "gemini-4-argon-high",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max"
        ],
        "id": "muse-wonder-max"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-5-pro",
          "mimo-v2-6-flash",
          "qwen3-7-plus",
          "spark-x2-6"
        ],
        "id": "nemotron-3-ultra"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3-1-flash",
          "qwen3-8-flash-next"
        ],
        "id": "qwen3-7-max"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "mimo-v2-6-flash"
        ],
        "id": "qwen3-7-plus"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "minimax-m3-1-flash"
        ],
        "id": "qwen3-8-27b"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "glm-5-3-flash",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro"
        ],
        "id": "qwen3-8-flash-next"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "claude-sonnet-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gemini-3-8-flash-high",
          "gemini-4-argon-high",
          "glm-5-2-max",
          "glm-5-3-max",
          "gpt-6-1-sol-max",
          "gpt-6-1-terra-max",
          "gpt-6-luna-max",
          "hy4-preview",
          "kimi-k3-max",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max"
        ],
        "id": "qwen3-8-max"
      },
      {
        "dominated_by": [
          "gpt-6-luna-low",
          "gpt-6-luna-medium"
        ],
        "id": "solar-pro-4"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "gpt-6-luna-low",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy3",
          "mimo-v2-5-pro",
          "mimo-v2-6-flash"
        ],
        "id": "spark-x2-6"
      },
      {
        "dominated_by": [
          "claude-haiku-5-5-high",
          "deepseek-v4-1-flash-20261017-max",
          "gemini-3-7-flash-high",
          "glm-5-3-flash",
          "gpt-5-6-terra-xhigh",
          "gpt-6-luna-max",
          "gpt-6-luna-medium",
          "hy4-preview",
          "mimo-v2-6-flash",
          "mimo-v2-6-pro",
          "muse-spark-1-3-max"
        ],
        "id": "step-5-preview"
      }
    ],
    "excluded": [
      {
        "id": "beam-1-0-501b-a23b",
        "reason": "unknown_cost"
      },
      {
        "id": "deepseek-v4-pro-high-20260813",
        "reason": "unknown_cost"
      },
      {
        "id": "muse-wonder-full-max",
        "reason": "unknown_cost"
      },
      {
        "id": "nvidia-nemotron-3-5-super-vl-120b-a12b-nvfp4",
        "reason": "unknown_cost"
      }
    ],
    "frontier": [
      "gpt-6-luna-low",
      "gpt-6-luna-medium",
      "mimo-v2-6-flash",
      "gpt-6-luna-max",
      "mimo-v2-6-pro",
      "muse-spark-1-3-max",
      "gemini-3-8-flash-high",
      "gpt-6-1-terra-max",
      "gpt-6-1-sol-max",
      "gemini-4-carbon-high",
      "claude-opus-5-5-max"
    ]
  },
  "prices": [
    {
      "confirmed": true,
      "cost": 6.81891,
      "plan_id": "arena-agent.claude-opus-5-5-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "claude-opus-5-5-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.733506,
      "plan_id": "arena-agent.gemini-4-carbon-high.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gemini-4-carbon-high",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 3.940786,
      "plan_id": "arena-agent.claude-sonnet-5-5-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "claude-sonnet-5-5-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 6.09363,
      "plan_id": "arena-agent.gpt-6-astra-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-6-astra-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 6.95462,
      "plan_id": "arena-agent.claude-fable-5-1-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "claude-fable-5-1-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.955171,
      "plan_id": "arena-agent.gpt-6-1-sol-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-6-1-sol-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.999277,
      "plan_id": "arena-agent.gemini-4-argon-high.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gemini-4-argon-high",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.787657,
      "plan_id": "arena-agent.gpt-6-1-terra-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-6-1-terra-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 3.917472,
      "plan_id": "arena-agent.claude-fable-5-high.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "claude-fable-5-high",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 5.624456,
      "plan_id": "arena-agent.claude-opus-5-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "claude-opus-5-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.738568,
      "plan_id": "arena-agent.gpt-6-sol-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-6-sol-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.467652,
      "plan_id": "arena-agent.kimi-k3-1-pro-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "kimi-k3-1-pro-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.728686,
      "plan_id": "arena-agent.gpt-5-6-sol-xhigh.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-5-6-sol-xhigh",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.112702,
      "plan_id": "arena-agent.claude-sonnet-5-high.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "claude-sonnet-5-high",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 2.611359,
      "plan_id": "arena-agent.grok-4-7-xhigh.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "grok-4-7-xhigh",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.757911,
      "plan_id": "arena-agent.gemini-3-8-flash-high.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gemini-3-8-flash-high",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.432282,
      "plan_id": "arena-agent.muse-wonder-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "muse-wonder-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.449172,
      "plan_id": "arena-agent.muse-spark-1-3-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "muse-spark-1-3-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.155964,
      "plan_id": "arena-agent.mimo-v2-6-pro.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "mimo-v2-6-pro",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.176969,
      "plan_id": "arena-agent.deepseek-v4-1-flash-20261017-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "deepseek-v4-1-flash-20261017-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.091385,
      "plan_id": "arena-agent.gpt-6-luna-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-6-luna-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.140962,
      "plan_id": "arena-agent.glm-5-3-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "glm-5-3-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.444307,
      "plan_id": "arena-agent.hy4-preview.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "hy4-preview",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.123475,
      "plan_id": "arena-agent.kimi-k3-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "kimi-k3-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.090219,
      "plan_id": "arena-agent.glm-5-2-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "glm-5-2-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.096408,
      "plan_id": "arena-agent.claude-haiku-5-5-high.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "claude-haiku-5-5-high",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.071299,
      "plan_id": "arena-agent.mimo-v2-6-flash.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "mimo-v2-6-flash",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.352575,
      "plan_id": "arena-agent.qwen3-8-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "qwen3-8-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 1.688277,
      "plan_id": "arena-agent.grok-4-6-xhigh.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "grok-4-6-xhigh",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.606392,
      "plan_id": "arena-agent.gpt-5-6-terra-xhigh.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-5-6-terra-xhigh",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.588672,
      "plan_id": "arena-agent.gemini-3-7-flash-high.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gemini-3-7-flash-high",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.027016,
      "plan_id": "arena-agent.gpt-6-luna-medium.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-6-luna-medium",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.160127,
      "plan_id": "arena-agent.glm-5-3-flash.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "glm-5-3-flash",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.631197,
      "plan_id": "arena-agent.step-5-preview.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "step-5-preview",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.241642,
      "plan_id": "arena-agent.minimax-m3-1-flash.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "minimax-m3-1-flash",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.430487,
      "plan_id": "arena-agent.qwen3-8-27b.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "qwen3-8-27b",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.221456,
      "plan_id": "arena-agent.qwen3-8-flash-next.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "qwen3-8-flash-next",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.632693,
      "plan_id": "arena-agent.gemini-3-1-pro-preview.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gemini-3-1-pro-preview",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.010452,
      "plan_id": "arena-agent.gpt-6-luna-low.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "gpt-6-luna-low",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.445551,
      "plan_id": "arena-agent.mistral-large-4.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "mistral-large-4",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.353168,
      "plan_id": "arena-agent.qwen3-7-max.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "qwen3-7-max",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.14057,
      "plan_id": "arena-agent.qwen3-7-plus.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "qwen3-7-plus",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.073335,
      "plan_id": "arena-agent.hy3.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "hy3",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.372478,
      "plan_id": "arena-agent.muse-spark-1-1.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "muse-spark-1-1",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.438531,
      "plan_id": "arena-agent.muse-spark-1-2-xhigh.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "muse-spark-1-2-xhigh",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.076846,
      "plan_id": "arena-agent.mimo-v2-5-pro.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "mimo-v2-5-pro",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.097772,
      "plan_id": "arena-agent.spark-x2-6.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "spark-x2-6",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.23518,
      "plan_id": "arena-agent.minimax-m3.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "minimax-m3",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.469507,
      "plan_id": "arena-agent.mistral-medium-3-5.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "mistral-medium-3-5",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.149715,
      "plan_id": "arena-agent.nemotron-3-ultra.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "nemotron-3-ultra",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.332419,
      "plan_id": "arena-agent.inkling.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "inkling",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.18831,
      "plan_id": "arena-agent.inkling-small.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "inkling-small",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    },
    {
      "confirmed": true,
      "cost": 0.05355,
      "plan_id": "arena-agent.solar-pro-4.cost_per_task_p50.2026-11-02",
      "price_expired": false,
      "product": "solar-pro-4",
      "retrieved_at": "2026-11-03",
      "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset",
      "unit": "USD per task (median)"
    }
  ],
  "products": [
    {
      "cost": {
        "confirmed": true,
        "cost": 6.81891,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.claude-opus-5-5-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "claude-opus-5-5-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "claude-opus-5-5-max",
      "name": "Claude Opus 5.5 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 94.295206,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 85.847444,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 90.071325,
            "used": [
              "arena-agent.claude-opus-5-5-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.200357,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 8.447762
          }
        ],
        "complete": true,
        "high": 94.295206,
        "label": "ci95",
        "low": 85.847444,
        "missing": [],
        "q": 90.071325,
        "spread_unknown": []
      },
      "vendor": "anthropic"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.733506,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gemini-4-carbon-high.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gemini-4-carbon-high",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gemini-4-carbon-high",
      "name": "Gemini 4 Carbon (High)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 84.724259,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 81.501485,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 83.112872,
            "used": [
              "arena-agent.gemini-4-carbon-high.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.165564,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.222774
          }
        ],
        "complete": true,
        "high": 84.724259,
        "label": "ci95",
        "low": 81.501485,
        "missing": [],
        "q": 83.112872,
        "spread_unknown": []
      },
      "vendor": "google"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 3.940786,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.claude-sonnet-5-5-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "claude-sonnet-5-5-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "claude-sonnet-5-5-max",
      "name": "Claude Sonnet 5.5 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 84.801461,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 73.660594,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 79.231028,
            "used": [
              "arena-agent.claude-sonnet-5-5-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.146155,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 11.140867
          }
        ],
        "complete": true,
        "high": 84.801461,
        "label": "ci95",
        "low": 73.660594,
        "missing": [],
        "q": 79.231028,
        "spread_unknown": []
      },
      "vendor": "anthropic"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 6.09363,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-6-astra-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-6-astra-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-6-astra-max",
      "name": "GPT-6 Astra (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 81.382177,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 73.344332,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 77.363254,
            "used": [
              "arena-agent.gpt-6-astra-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.136816,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 8.037845
          }
        ],
        "complete": true,
        "high": 81.382177,
        "label": "ci95",
        "low": 73.344332,
        "missing": [],
        "q": 77.363254,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 6.95462,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.claude-fable-5-1-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "claude-fable-5-1-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "claude-fable-5-1-max",
      "name": "Claude Fable 5.1 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 80.047018,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 72.536559,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 76.291788,
            "used": [
              "arena-agent.claude-fable-5-1-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.131459,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 7.510459
          }
        ],
        "complete": true,
        "high": 80.047018,
        "label": "ci95",
        "low": 72.536559,
        "missing": [],
        "q": 76.291788,
        "spread_unknown": []
      },
      "vendor": "anthropic"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.955171,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-6-1-sol-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-6-1-sol-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-6-1-sol-max",
      "name": "GPT-6.1 Sol (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 76.660254,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 72.292235,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 74.476245,
            "used": [
              "arena-agent.gpt-6-1-sol-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.122381,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.368019
          }
        ],
        "complete": true,
        "high": 76.660254,
        "label": "ci95",
        "low": 72.292235,
        "missing": [],
        "q": 74.476245,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.999277,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gemini-4-argon-high.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gemini-4-argon-high",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gemini-4-argon-high",
      "name": "Gemini 4 Argon (High)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 72.907038,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 69.520859,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 71.213948,
            "used": [
              "arena-agent.gemini-4-argon-high.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.10607,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.386179
          }
        ],
        "complete": true,
        "high": 72.907038,
        "label": "ci95",
        "low": 69.520859,
        "missing": [],
        "q": 71.213948,
        "spread_unknown": []
      },
      "vendor": "google"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.787657,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-6-1-terra-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-6-1-terra-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-6-1-terra-max",
      "name": "GPT-6.1 Terra (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 72.699035,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 68.048199,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 70.373617,
            "used": [
              "arena-agent.gpt-6-1-terra-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.101868,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.650836
          }
        ],
        "complete": true,
        "high": 72.699035,
        "label": "ci95",
        "low": 68.048199,
        "missing": [],
        "q": 70.373617,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": false,
        "cost": null,
        "currency": null,
        "flags": [],
        "plan": null,
        "price_age_days": null,
        "price_expired": false,
        "product": "muse-wonder-full-max",
        "rejected": [],
        "status": "no_price",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "muse-wonder-full-max",
      "name": "Muse Wonder Full (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 71.243241,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 67.296257,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 69.269749,
            "used": [
              "arena-agent.muse-wonder-full-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.096349,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.946984
          }
        ],
        "complete": true,
        "high": 71.243241,
        "label": "ci95",
        "low": 67.296257,
        "missing": [],
        "q": 69.269749,
        "spread_unknown": []
      },
      "vendor": "meta"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 3.917472,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.claude-fable-5-high.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "claude-fable-5-high",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "claude-fable-5-high",
      "name": "Claude Fable 5 (High)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 67.781908,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 63.329136,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 65.555522,
            "used": [
              "arena-agent.claude-fable-5-high.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.077778,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.452772
          }
        ],
        "complete": true,
        "high": 67.781908,
        "label": "ci95",
        "low": 63.329136,
        "missing": [],
        "q": 65.555522,
        "spread_unknown": []
      },
      "vendor": "anthropic"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 5.624456,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.claude-opus-5-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "claude-opus-5-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "claude-opus-5-max",
      "name": "Claude Opus 5 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 67.679727,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 62.717029,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 65.198378,
            "used": [
              "arena-agent.claude-opus-5-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.075992,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.962698
          }
        ],
        "complete": true,
        "high": 67.679727,
        "label": "ci95",
        "low": 62.717029,
        "missing": [],
        "q": 65.198378,
        "spread_unknown": []
      },
      "vendor": "anthropic"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.738568,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-6-sol-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-6-sol-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-6-sol-max",
      "name": "GPT-6 Sol (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 67.072466,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 60.714244,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 63.893355,
            "used": [
              "arena-agent.gpt-6-sol-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.069467,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 6.358222
          }
        ],
        "complete": true,
        "high": 67.072466,
        "label": "ci95",
        "low": 60.714244,
        "missing": [],
        "q": 63.893355,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.467652,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.kimi-k3-1-pro-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "kimi-k3-1-pro-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "kimi-k3-1-pro-max",
      "name": "Kimi K3.1 Pro (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 62.033048,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 56.750669,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 59.391858,
            "used": [
              "arena-agent.kimi-k3-1-pro-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.046959,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.282379
          }
        ],
        "complete": true,
        "high": 62.033048,
        "label": "ci95",
        "low": 56.750669,
        "missing": [],
        "q": 59.391858,
        "spread_unknown": []
      },
      "vendor": "moonshot"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.728686,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-5-6-sol-xhigh.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-5-6-sol-xhigh",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-5-6-sol-xhigh",
      "name": "GPT-5.6 Sol (xHigh)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 61.195307,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 57.036953,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 59.11613,
            "used": [
              "arena-agent.gpt-5-6-sol-xhigh.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.045581,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.158354
          }
        ],
        "complete": true,
        "high": 61.195307,
        "label": "ci95",
        "low": 57.036953,
        "missing": [],
        "q": 59.11613,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.112702,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.claude-sonnet-5-high.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "claude-sonnet-5-high",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "claude-sonnet-5-high",
      "name": "Claude Sonnet 5 (High)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 58.05674,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 53.309652,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 55.683196,
            "used": [
              "arena-agent.claude-sonnet-5-high.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.028416,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.747088
          }
        ],
        "complete": true,
        "high": 58.05674,
        "label": "ci95",
        "low": 53.309652,
        "missing": [],
        "q": 55.683196,
        "spread_unknown": []
      },
      "vendor": "anthropic"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 2.611359,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.grok-4-7-xhigh.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "grok-4-7-xhigh",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "grok-4-7-xhigh",
      "name": "Grok 4.7 (xHigh)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 58.218349,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 52.355838,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 55.287093,
            "used": [
              "arena-agent.grok-4-7-xhigh.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.026435,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.862511
          }
        ],
        "complete": true,
        "high": 58.218349,
        "label": "ci95",
        "low": 52.355838,
        "missing": [],
        "q": 55.287093,
        "spread_unknown": []
      },
      "vendor": "xai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.757911,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gemini-3-8-flash-high.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gemini-3-8-flash-high",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gemini-3-8-flash-high",
      "name": "Gemini 3.8 Flash (High)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 56.768598,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 53.476647,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 55.122623,
            "used": [
              "arena-agent.gemini-3-8-flash-high.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.025613,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.291951
          }
        ],
        "complete": true,
        "high": 56.768598,
        "label": "ci95",
        "low": 53.476647,
        "missing": [],
        "q": 55.122623,
        "spread_unknown": []
      },
      "vendor": "google"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.432282,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.muse-wonder-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "muse-wonder-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "muse-wonder-max",
      "name": "Muse Wonder (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 57.753098,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 52.289877,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 55.021488,
            "used": [
              "arena-agent.muse-wonder-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.025107,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.463221
          }
        ],
        "complete": true,
        "high": 57.753098,
        "label": "ci95",
        "low": 52.289877,
        "missing": [],
        "q": 55.021488,
        "spread_unknown": []
      },
      "vendor": "meta"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.449172,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.muse-spark-1-3-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "muse-spark-1-3-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "muse-spark-1-3-max",
      "name": "Muse Spark 1.3 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 56.528862,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 53.239007,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 54.883934,
            "used": [
              "arena-agent.muse-spark-1-3-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.02442,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.289855
          }
        ],
        "complete": true,
        "high": 56.528862,
        "label": "ci95",
        "low": 53.239007,
        "missing": [],
        "q": 54.883934,
        "spread_unknown": []
      },
      "vendor": "meta"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.155964,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.mimo-v2-6-pro.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "mimo-v2-6-pro",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "mimo-v2-6-pro",
      "name": "MiMo-V2.6-Pro",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 56.695462,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 53.070238,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 54.88285,
            "used": [
              "arena-agent.mimo-v2-6-pro.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.024414,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.625224
          }
        ],
        "complete": true,
        "high": 56.695462,
        "label": "ci95",
        "low": 53.070238,
        "missing": [],
        "q": 54.88285,
        "spread_unknown": []
      },
      "vendor": "xiaomi"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.176969,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.deepseek-v4-1-flash-20261017-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "deepseek-v4-1-flash-20261017-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "deepseek-v4-1-flash-20261017-max",
      "name": "DeepSeek-V4.1-Flash-20261017 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 55.422486,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 53.1017,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 54.262093,
            "used": [
              "arena-agent.deepseek-v4-1-flash-20261017-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.02131,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.320786
          }
        ],
        "complete": true,
        "high": 55.422486,
        "label": "ci95",
        "low": 53.1017,
        "missing": [],
        "q": 54.262093,
        "spread_unknown": []
      },
      "vendor": "deepseek"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.091385,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-6-luna-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-6-luna-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-6-luna-max",
      "name": "GPT-6 Luna (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 54.686685,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 52.979063,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 53.832874,
            "used": [
              "arena-agent.gpt-6-luna-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.019164,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 1.707622
          }
        ],
        "complete": true,
        "high": 54.686685,
        "label": "ci95",
        "low": 52.979063,
        "missing": [],
        "q": 53.832874,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.140962,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.glm-5-3-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "glm-5-3-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "glm-5-3-max",
      "name": "GLM-5.3 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 53.916908,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 50.745042,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 52.330975,
            "used": [
              "arena-agent.glm-5-3-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.011655,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.171866
          }
        ],
        "complete": true,
        "high": 53.916908,
        "label": "ci95",
        "low": 50.745042,
        "missing": [],
        "q": 52.330975,
        "spread_unknown": []
      },
      "vendor": "zai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.444307,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.hy4-preview.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "hy4-preview",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "hy4-preview",
      "name": "Hy4 preview",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 53.265591,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 50.482115,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 51.873853,
            "used": [
              "arena-agent.hy4-preview.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.009369,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.783476
          }
        ],
        "complete": true,
        "high": 53.265591,
        "label": "ci95",
        "low": 50.482115,
        "missing": [],
        "q": 51.873853,
        "spread_unknown": []
      },
      "vendor": "tencent"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.123475,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.kimi-k3-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "kimi-k3-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "kimi-k3-max",
      "name": "Kimi K3 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 52.858745,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 50.561416,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 51.710081,
            "used": [
              "arena-agent.kimi-k3-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.00855,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.297329
          }
        ],
        "complete": true,
        "high": 52.858745,
        "label": "ci95",
        "low": 50.561416,
        "missing": [],
        "q": 51.710081,
        "spread_unknown": []
      },
      "vendor": "moonshot"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.090219,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.glm-5-2-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "glm-5-2-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "glm-5-2-max",
      "name": "GLM-5.2 (Max)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 53.023652,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 49.959638,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 51.491645,
            "used": [
              "arena-agent.glm-5-2-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.007458,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.064014
          }
        ],
        "complete": true,
        "high": 53.023652,
        "label": "ci95",
        "low": 49.959638,
        "missing": [],
        "q": 51.491645,
        "spread_unknown": []
      },
      "vendor": "zai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.096408,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.claude-haiku-5-5-high.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "claude-haiku-5-5-high",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "claude-haiku-5-5-high",
      "name": "Claude Haiku 5.5 (High)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 53.364618,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 49.251697,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 51.308158,
            "used": [
              "arena-agent.claude-haiku-5-5-high.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.006541,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.112921
          }
        ],
        "complete": true,
        "high": 53.364618,
        "label": "ci95",
        "low": 49.251697,
        "missing": [],
        "q": 51.308158,
        "spread_unknown": []
      },
      "vendor": "anthropic"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.071299,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.mimo-v2-6-flash.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "mimo-v2-6-flash",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "mimo-v2-6-flash",
      "name": "MiMo-V2.6-Flash",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 52.954187,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 48.770118,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 50.862152,
            "used": [
              "arena-agent.mimo-v2-6-flash.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.004311,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.184069
          }
        ],
        "complete": true,
        "high": 52.954187,
        "label": "ci95",
        "low": 48.770118,
        "missing": [],
        "q": 50.862152,
        "spread_unknown": []
      },
      "vendor": "xiaomi"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.352575,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.qwen3-8-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "qwen3-8-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "qwen3-8-max",
      "name": "Qwen3.8-Max",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 52.103495,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 49.511261,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 50.807378,
            "used": [
              "arena-agent.qwen3-8-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.004037,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.592234
          }
        ],
        "complete": true,
        "high": 52.103495,
        "label": "ci95",
        "low": 49.511261,
        "missing": [],
        "q": 50.807378,
        "spread_unknown": []
      },
      "vendor": "alibaba"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 1.688277,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.grok-4-6-xhigh.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "grok-4-6-xhigh",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "grok-4-6-xhigh",
      "name": "Grok 4.6 (xHigh)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 52.938232,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 48.360148,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 50.64919,
            "used": [
              "arena-agent.grok-4-6-xhigh.agent_arena_ips_score.2026-11-02"
            ],
            "value": 0.003246,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.578084
          }
        ],
        "complete": true,
        "high": 52.938232,
        "label": "ci95",
        "low": 48.360148,
        "missing": [],
        "q": 50.64919,
        "spread_unknown": []
      },
      "vendor": "xai"
    },
    {
      "cost": {
        "confirmed": false,
        "cost": null,
        "currency": null,
        "flags": [],
        "plan": null,
        "price_age_days": null,
        "price_expired": false,
        "product": "deepseek-v4-pro-high-20260813",
        "rejected": [],
        "status": "no_price",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "deepseek-v4-pro-high-20260813",
      "name": "DeepSeek-V4-Pro (High, 20260813)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 52.802401,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 45.994168,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 49.398284,
            "used": [
              "arena-agent.deepseek-v4-pro-high-20260813.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.003009,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 6.808233
          }
        ],
        "complete": true,
        "high": 52.802401,
        "label": "ci95",
        "low": 45.994168,
        "missing": [],
        "q": 49.398284,
        "spread_unknown": []
      },
      "vendor": "deepseek"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.606392,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-5-6-terra-xhigh.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-5-6-terra-xhigh",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-5-6-terra-xhigh",
      "name": "GPT-5.6 Terra (xHigh)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 48.534184,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 45.415967,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 46.975075,
            "used": [
              "arena-agent.gpt-5-6-terra-xhigh.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.015125,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.118217
          }
        ],
        "complete": true,
        "high": 48.534184,
        "label": "ci95",
        "low": 45.415967,
        "missing": [],
        "q": 46.975075,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.588672,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gemini-3-7-flash-high.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gemini-3-7-flash-high",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gemini-3-7-flash-high",
      "name": "Gemini 3.7 Flash (High)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 48.486237,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 45.010029,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 46.748133,
            "used": [
              "arena-agent.gemini-3-7-flash-high.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.016259,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.476208
          }
        ],
        "complete": true,
        "high": 48.486237,
        "label": "ci95",
        "low": 45.010029,
        "missing": [],
        "q": 46.748133,
        "spread_unknown": []
      },
      "vendor": "google"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.027016,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-6-luna-medium.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-6-luna-medium",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-6-luna-medium",
      "name": "GPT-6 Luna (Medium)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 47.266981,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 45.014132,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 46.140557,
            "used": [
              "arena-agent.gpt-6-luna-medium.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.019297,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.252849
          }
        ],
        "complete": true,
        "high": 47.266981,
        "label": "ci95",
        "low": 45.014132,
        "missing": [],
        "q": 46.140557,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.160127,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.glm-5-3-flash.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "glm-5-3-flash",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "glm-5-3-flash",
      "name": "GLM-5.3-Flash",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 46.393353,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 43.64741,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 45.020381,
            "used": [
              "arena-agent.glm-5-3-flash.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.024898,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.745943
          }
        ],
        "complete": true,
        "high": 46.393353,
        "label": "ci95",
        "low": 43.64741,
        "missing": [],
        "q": 45.020381,
        "spread_unknown": []
      },
      "vendor": "zai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.631197,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.step-5-preview.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "step-5-preview",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "step-5-preview",
      "name": "Step 5 Preview",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 47.734756,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 42.134841,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 44.934799,
            "used": [
              "arena-agent.step-5-preview.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.025326,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.599915
          }
        ],
        "complete": true,
        "high": 47.734756,
        "label": "ci95",
        "low": 42.134841,
        "missing": [],
        "q": 44.934799,
        "spread_unknown": []
      },
      "vendor": "stepfun"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.241642,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.minimax-m3-1-flash.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "minimax-m3-1-flash",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "minimax-m3-1-flash",
      "name": "MiniMax-M3.1-Flash",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 46.693714,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 40.440989,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 43.567351,
            "used": [
              "arena-agent.minimax-m3-1-flash.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.032163,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 6.252725
          }
        ],
        "complete": true,
        "high": 46.693714,
        "label": "ci95",
        "low": 40.440989,
        "missing": [],
        "q": 43.567351,
        "spread_unknown": []
      },
      "vendor": "minimax"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.430487,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.qwen3-8-27b.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "qwen3-8-27b",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "qwen3-8-27b",
      "name": "Qwen3.8-27B",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 43.084554,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 39.7127,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 41.398627,
            "used": [
              "arena-agent.qwen3-8-27b.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.043007,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.371854
          }
        ],
        "complete": true,
        "high": 43.084554,
        "label": "ci95",
        "low": 39.7127,
        "missing": [],
        "q": 41.398627,
        "spread_unknown": []
      },
      "vendor": "alibaba"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.221456,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.qwen3-8-flash-next.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "qwen3-8-flash-next",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "qwen3-8-flash-next",
      "name": "Qwen3.8-Flash-Next",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 42.779762,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 39.884921,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 41.332341,
            "used": [
              "arena-agent.qwen3-8-flash-next.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.043338,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.894841
          }
        ],
        "complete": true,
        "high": 42.779762,
        "label": "ci95",
        "low": 39.884921,
        "missing": [],
        "q": 41.332341,
        "spread_unknown": []
      },
      "vendor": "alibaba"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.632693,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gemini-3-1-pro-preview.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gemini-3-1-pro-preview",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gemini-3-1-pro-preview",
      "name": "Gemini 3.1 Pro Preview",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 43.535596,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 38.006536,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 40.771066,
            "used": [
              "arena-agent.gemini-3-1-pro-preview.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.046145,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.52906
          }
        ],
        "complete": true,
        "high": 43.535596,
        "label": "ci95",
        "low": 38.006536,
        "missing": [],
        "q": 40.771066,
        "spread_unknown": []
      },
      "vendor": "google"
    },
    {
      "cost": {
        "confirmed": false,
        "cost": null,
        "currency": null,
        "flags": [],
        "plan": null,
        "price_age_days": null,
        "price_expired": false,
        "product": "beam-1-0-501b-a23b",
        "rejected": [],
        "status": "no_price",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "beam-1-0-501b-a23b",
      "name": "Beam 1.0 501B A23B",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 45.736055,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 35.478759,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 40.607407,
            "used": [
              "arena-agent.beam-1-0-501b-a23b.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.046963,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 10.257296
          }
        ],
        "complete": true,
        "high": 45.736055,
        "label": "ci95",
        "low": 35.478759,
        "missing": [],
        "q": 40.607407,
        "spread_unknown": []
      },
      "vendor": "reflection"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.010452,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.gpt-6-luna-low.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "gpt-6-luna-low",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "gpt-6-luna-low",
      "name": "GPT-6 Luna (Low)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 39.719359,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 36.79189,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 38.255624,
            "used": [
              "arena-agent.gpt-6-luna-low.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.058722,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.927469
          }
        ],
        "complete": true,
        "high": 39.719359,
        "label": "ci95",
        "low": 36.79189,
        "missing": [],
        "q": 38.255624,
        "spread_unknown": []
      },
      "vendor": "openai"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.445551,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.mistral-large-4.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "mistral-large-4",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "mistral-large-4",
      "name": "Mistral Large 4",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 41.407183,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 32.954264,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 37.180723,
            "used": [
              "arena-agent.mistral-large-4.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.064096,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 8.452919
          }
        ],
        "complete": true,
        "high": 41.407183,
        "label": "ci95",
        "low": 32.954264,
        "missing": [],
        "q": 37.180723,
        "spread_unknown": []
      },
      "vendor": "mistral"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.353168,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.qwen3-7-max.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "qwen3-7-max",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "qwen3-7-max",
      "name": "Qwen3.7-Max",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 39.058923,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 34.263497,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 36.66121,
            "used": [
              "arena-agent.qwen3-7-max.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.066694,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.795426
          }
        ],
        "complete": true,
        "high": 39.058923,
        "label": "ci95",
        "low": 34.263497,
        "missing": [],
        "q": 36.66121,
        "spread_unknown": []
      },
      "vendor": "alibaba"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.14057,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.qwen3-7-plus.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "qwen3-7-plus",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "qwen3-7-plus",
      "name": "Qwen3.7-Plus",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 38.499268,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 32.03128,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 35.265274,
            "used": [
              "arena-agent.qwen3-7-plus.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.073674,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 6.467988
          }
        ],
        "complete": true,
        "high": 38.499268,
        "label": "ci95",
        "low": 32.03128,
        "missing": [],
        "q": 35.265274,
        "spread_unknown": []
      },
      "vendor": "alibaba"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.073335,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.hy3.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "hy3",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "hy3",
      "name": "Hy3",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 37.807241,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 32.054089,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 34.930665,
            "used": [
              "arena-agent.hy3.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.075347,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.753152
          }
        ],
        "complete": true,
        "high": 37.807241,
        "label": "ci95",
        "low": 32.054089,
        "missing": [],
        "q": 34.930665,
        "spread_unknown": []
      },
      "vendor": "tencent"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.372478,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.muse-spark-1-1.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "muse-spark-1-1",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "muse-spark-1-1",
      "name": "Muse Spark 1.1",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 35.966836,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 33.33192,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 34.649378,
            "used": [
              "arena-agent.muse-spark-1-1.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.076753,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 2.634916
          }
        ],
        "complete": true,
        "high": 35.966836,
        "label": "ci95",
        "low": 33.33192,
        "missing": [],
        "q": 34.649378,
        "spread_unknown": []
      },
      "vendor": "meta"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.438531,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.muse-spark-1-2-xhigh.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "muse-spark-1-2-xhigh",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "muse-spark-1-2-xhigh",
      "name": "Muse Spark 1.2 (xHigh)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 36.201892,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 32.832457,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 34.517175,
            "used": [
              "arena-agent.muse-spark-1-2-xhigh.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.077414,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.369435
          }
        ],
        "complete": true,
        "high": 36.201892,
        "label": "ci95",
        "low": 32.832457,
        "missing": [],
        "q": 34.517175,
        "spread_unknown": []
      },
      "vendor": "meta"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.076846,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.mimo-v2-5-pro.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "mimo-v2-5-pro",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "mimo-v2-5-pro",
      "name": "MiMo-V2.5-Pro",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 34.003631,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 29.949232,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 31.976431,
            "used": [
              "arena-agent.mimo-v2-5-pro.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.090118,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.054399
          }
        ],
        "complete": true,
        "high": 34.003631,
        "label": "ci95",
        "low": 29.949232,
        "missing": [],
        "q": 31.976431,
        "spread_unknown": []
      },
      "vendor": "xiaomi"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.097772,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.spark-x2-6.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "spark-x2-6",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "spark-x2-6",
      "name": "Spark-X2.6",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 32.97231,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 27.01375,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 29.99303,
            "used": [
              "arena-agent.spark-x2-6.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.100035,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.95856
          }
        ],
        "complete": true,
        "high": 32.97231,
        "label": "ci95",
        "low": 27.01375,
        "missing": [],
        "q": 29.99303,
        "spread_unknown": []
      },
      "vendor": "synlan"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.23518,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.minimax-m3.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "minimax-m3",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "minimax-m3",
      "name": "MiniMax-M3",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 31.516786,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 26.82212,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 29.169453,
            "used": [
              "arena-agent.minimax-m3.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.104153,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 4.694666
          }
        ],
        "complete": true,
        "high": 31.516786,
        "label": "ci95",
        "low": 26.82212,
        "missing": [],
        "q": 29.169453,
        "spread_unknown": []
      },
      "vendor": "minimax"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.469507,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.mistral-medium-3-5.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "mistral-medium-3-5",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "mistral-medium-3-5",
      "name": "Mistral Medium 3.5",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 26.960364,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 20.174465,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 23.567415,
            "used": [
              "arena-agent.mistral-medium-3-5.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.132163,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 6.785899
          }
        ],
        "complete": true,
        "high": 26.960364,
        "label": "ci95",
        "low": 20.174465,
        "missing": [],
        "q": 23.567415,
        "spread_unknown": []
      },
      "vendor": "mistral"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.149715,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.nemotron-3-ultra.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "nemotron-3-ultra",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "nemotron-3-ultra",
      "name": "Nemotron 3 Ultra",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 25.229886,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 12.363133,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 18.796509,
            "used": [
              "arena-agent.nemotron-3-ultra.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.156017,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 12.866753
          }
        ],
        "complete": true,
        "high": 25.229886,
        "label": "ci95",
        "low": 12.363133,
        "missing": [],
        "q": 18.796509,
        "spread_unknown": []
      },
      "vendor": "nvidia"
    },
    {
      "cost": {
        "confirmed": false,
        "cost": null,
        "currency": null,
        "flags": [],
        "plan": null,
        "price_age_days": null,
        "price_expired": false,
        "product": "nvidia-nemotron-3-5-super-vl-120b-a12b-nvfp4",
        "rejected": [],
        "status": "no_price",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "nvidia-nemotron-3-5-super-vl-120b-a12b-nvfp4",
      "name": "NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4)",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 20.11268,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 16.226446,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 18.169563,
            "used": [
              "arena-agent.nvidia-nemotron-3-5-super-vl-120b-a12b-nvfp4.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.159152,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 3.886234
          }
        ],
        "complete": true,
        "high": 20.11268,
        "label": "ci95",
        "low": 16.226446,
        "missing": [],
        "q": 18.169563,
        "spread_unknown": []
      },
      "vendor": "nvidia"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.332419,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.inkling.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "inkling",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "inkling",
      "name": "Inkling",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 20.801526,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 14.864282,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 17.832904,
            "used": [
              "arena-agent.inkling.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.160835,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 5.937244
          }
        ],
        "complete": true,
        "high": 20.801526,
        "label": "ci95",
        "low": 14.864282,
        "missing": [],
        "q": 17.832904,
        "spread_unknown": []
      },
      "vendor": "thinky"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.18831,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.inkling-small.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "inkling-small",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "inkling-small",
      "name": "Inkling-Small",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 19.948368,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 12.485999,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 16.217183,
            "used": [
              "arena-agent.inkling-small.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.168914,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 7.462369
          }
        ],
        "complete": true,
        "high": 19.948368,
        "label": "ci95",
        "low": 12.485999,
        "missing": [],
        "q": 16.217183,
        "spread_unknown": []
      },
      "vendor": "thinky"
    },
    {
      "cost": {
        "confirmed": true,
        "cost": 0.05355,
        "currency": null,
        "flags": [],
        "plan": {
          "id": "arena-agent.solar-pro-4.cost_per_task_p50.2026-11-02",
          "retrieved_at": "2026-11-03",
          "source_id": "arena-leaderboard-dataset",
          "source_url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
        },
        "price_age_days": 2,
        "price_expired": false,
        "product": "solar-pro-4",
        "rejected": [],
        "status": "confirmed",
        "unconfirmed_plans": [],
        "unit": "USD per task (median)",
        "workload_chars": null
      },
      "id": "solar-pro-4",
      "name": "Solar Pro 4",
      "quality": {
        "breakdown": [
          {
            "auditable_own": false,
            "conflicts": [
              "unknown"
            ],
            "critical": true,
            "excluded": [],
            "high": 12.308342,
            "key": "agent_arena_ips_score",
            "kind": "ci95",
            "low": 3.912338,
            "missing_reason": null,
            "origins": 1,
            "present": true,
            "u_hat": 8.11034,
            "used": [
              "arena-agent.solar-pro-4.agent_arena_ips_score.2026-11-02"
            ],
            "value": -0.209448,
            "weight": 1,
            "weight_total": 0.466516,
            "width_contribution": 8.396004
          }
        ],
        "complete": true,
        "high": 12.308342,
        "label": "ci95",
        "low": 3.912338,
        "missing": [],
        "q": 8.11034,
        "spread_unknown": []
      },
      "vendor": "upstage"
    }
  ],
  "ranking": [
    {
      "cost": 6.81891,
      "cost_confirmed": true,
      "high": 94.295206,
      "label": "ci95",
      "low": 85.847444,
      "product": "claude-opus-5-5-max",
      "q": 90.071325,
      "rank": 1,
      "status": "active"
    },
    {
      "cost": 1.733506,
      "cost_confirmed": true,
      "high": 84.724259,
      "label": "ci95",
      "low": 81.501485,
      "product": "gemini-4-carbon-high",
      "q": 83.112872,
      "rank": 2,
      "status": "active"
    },
    {
      "cost": 3.940786,
      "cost_confirmed": true,
      "high": 84.801461,
      "label": "ci95",
      "low": 73.660594,
      "product": "claude-sonnet-5-5-max",
      "q": 79.231028,
      "rank": 3,
      "status": "active"
    },
    {
      "cost": 6.09363,
      "cost_confirmed": true,
      "high": 81.382177,
      "label": "ci95",
      "low": 73.344332,
      "product": "gpt-6-astra-max",
      "q": 77.363254,
      "rank": 4,
      "status": "active"
    },
    {
      "cost": 6.95462,
      "cost_confirmed": true,
      "high": 80.047018,
      "label": "ci95",
      "low": 72.536559,
      "product": "claude-fable-5-1-max",
      "q": 76.291788,
      "rank": 5,
      "status": "active"
    },
    {
      "cost": 0.955171,
      "cost_confirmed": true,
      "high": 76.660254,
      "label": "ci95",
      "low": 72.292235,
      "product": "gpt-6-1-sol-max",
      "q": 74.476245,
      "rank": 6,
      "status": "active"
    },
    {
      "cost": 0.999277,
      "cost_confirmed": true,
      "high": 72.907038,
      "label": "ci95",
      "low": 69.520859,
      "product": "gemini-4-argon-high",
      "q": 71.213948,
      "rank": 7,
      "status": "active"
    },
    {
      "cost": 0.787657,
      "cost_confirmed": true,
      "high": 72.699035,
      "label": "ci95",
      "low": 68.048199,
      "product": "gpt-6-1-terra-max",
      "q": 70.373617,
      "rank": 8,
      "status": "active"
    },
    {
      "cost": null,
      "cost_confirmed": false,
      "high": 71.243241,
      "label": "ci95",
      "low": 67.296257,
      "product": "muse-wonder-full-max",
      "q": 69.269749,
      "rank": 9,
      "status": "active"
    },
    {
      "cost": 3.917472,
      "cost_confirmed": true,
      "high": 67.781908,
      "label": "ci95",
      "low": 63.329136,
      "product": "claude-fable-5-high",
      "q": 65.555522,
      "rank": 10,
      "status": "active"
    },
    {
      "cost": 5.624456,
      "cost_confirmed": true,
      "high": 67.679727,
      "label": "ci95",
      "low": 62.717029,
      "product": "claude-opus-5-max",
      "q": 65.198378,
      "rank": 11,
      "status": "active"
    },
    {
      "cost": 1.738568,
      "cost_confirmed": true,
      "high": 67.072466,
      "label": "ci95",
      "low": 60.714244,
      "product": "gpt-6-sol-max",
      "q": 63.893355,
      "rank": 12,
      "status": "active"
    },
    {
      "cost": 1.467652,
      "cost_confirmed": true,
      "high": 62.033048,
      "label": "ci95",
      "low": 56.750669,
      "product": "kimi-k3-1-pro-max",
      "q": 59.391858,
      "rank": 13,
      "status": "active"
    },
    {
      "cost": 1.728686,
      "cost_confirmed": true,
      "high": 61.195307,
      "label": "ci95",
      "low": 57.036953,
      "product": "gpt-5-6-sol-xhigh",
      "q": 59.11613,
      "rank": 14,
      "status": "active"
    },
    {
      "cost": 1.112702,
      "cost_confirmed": true,
      "high": 58.05674,
      "label": "ci95",
      "low": 53.309652,
      "product": "claude-sonnet-5-high",
      "q": 55.683196,
      "rank": 15,
      "status": "active"
    },
    {
      "cost": 2.611359,
      "cost_confirmed": true,
      "high": 58.218349,
      "label": "ci95",
      "low": 52.355838,
      "product": "grok-4-7-xhigh",
      "q": 55.287093,
      "rank": 16,
      "status": "active"
    },
    {
      "cost": 0.757911,
      "cost_confirmed": true,
      "high": 56.768598,
      "label": "ci95",
      "low": 53.476647,
      "product": "gemini-3-8-flash-high",
      "q": 55.122623,
      "rank": 17,
      "status": "active"
    },
    {
      "cost": 1.432282,
      "cost_confirmed": true,
      "high": 57.753098,
      "label": "ci95",
      "low": 52.289877,
      "product": "muse-wonder-max",
      "q": 55.021488,
      "rank": 18,
      "status": "active"
    },
    {
      "cost": 0.449172,
      "cost_confirmed": true,
      "high": 56.528862,
      "label": "ci95",
      "low": 53.239007,
      "product": "muse-spark-1-3-max",
      "q": 54.883934,
      "rank": 19,
      "status": "active"
    },
    {
      "cost": 0.155964,
      "cost_confirmed": true,
      "high": 56.695462,
      "label": "ci95",
      "low": 53.070238,
      "product": "mimo-v2-6-pro",
      "q": 54.88285,
      "rank": 20,
      "status": "active"
    },
    {
      "cost": 0.176969,
      "cost_confirmed": true,
      "high": 55.422486,
      "label": "ci95",
      "low": 53.1017,
      "product": "deepseek-v4-1-flash-20261017-max",
      "q": 54.262093,
      "rank": 21,
      "status": "active"
    },
    {
      "cost": 0.091385,
      "cost_confirmed": true,
      "high": 54.686685,
      "label": "ci95",
      "low": 52.979063,
      "product": "gpt-6-luna-max",
      "q": 53.832874,
      "rank": 22,
      "status": "active"
    },
    {
      "cost": 1.140962,
      "cost_confirmed": true,
      "high": 53.916908,
      "label": "ci95",
      "low": 50.745042,
      "product": "glm-5-3-max",
      "q": 52.330975,
      "rank": 23,
      "status": "active"
    },
    {
      "cost": 0.444307,
      "cost_confirmed": true,
      "high": 53.265591,
      "label": "ci95",
      "low": 50.482115,
      "product": "hy4-preview",
      "q": 51.873853,
      "rank": 24,
      "status": "active"
    },
    {
      "cost": 1.123475,
      "cost_confirmed": true,
      "high": 52.858745,
      "label": "ci95",
      "low": 50.561416,
      "product": "kimi-k3-max",
      "q": 51.710081,
      "rank": 25,
      "status": "active"
    },
    {
      "cost": 1.090219,
      "cost_confirmed": true,
      "high": 53.023652,
      "label": "ci95",
      "low": 49.959638,
      "product": "glm-5-2-max",
      "q": 51.491645,
      "rank": 26,
      "status": "active"
    },
    {
      "cost": 0.096408,
      "cost_confirmed": true,
      "high": 53.364618,
      "label": "ci95",
      "low": 49.251697,
      "product": "claude-haiku-5-5-high",
      "q": 51.308158,
      "rank": 27,
      "status": "active"
    },
    {
      "cost": 0.071299,
      "cost_confirmed": true,
      "high": 52.954187,
      "label": "ci95",
      "low": 48.770118,
      "product": "mimo-v2-6-flash",
      "q": 50.862152,
      "rank": 28,
      "status": "active"
    },
    {
      "cost": 1.352575,
      "cost_confirmed": true,
      "high": 52.103495,
      "label": "ci95",
      "low": 49.511261,
      "product": "qwen3-8-max",
      "q": 50.807378,
      "rank": 29,
      "status": "active"
    },
    {
      "cost": 1.688277,
      "cost_confirmed": true,
      "high": 52.938232,
      "label": "ci95",
      "low": 48.360148,
      "product": "grok-4-6-xhigh",
      "q": 50.64919,
      "rank": 30,
      "status": "active"
    },
    {
      "cost": null,
      "cost_confirmed": false,
      "high": 52.802401,
      "label": "ci95",
      "low": 45.994168,
      "product": "deepseek-v4-pro-high-20260813",
      "q": 49.398284,
      "rank": 31,
      "status": "active"
    },
    {
      "cost": 0.606392,
      "cost_confirmed": true,
      "high": 48.534184,
      "label": "ci95",
      "low": 45.415967,
      "product": "gpt-5-6-terra-xhigh",
      "q": 46.975075,
      "rank": 32,
      "status": "active"
    },
    {
      "cost": 0.588672,
      "cost_confirmed": true,
      "high": 48.486237,
      "label": "ci95",
      "low": 45.010029,
      "product": "gemini-3-7-flash-high",
      "q": 46.748133,
      "rank": 33,
      "status": "active"
    },
    {
      "cost": 0.027016,
      "cost_confirmed": true,
      "high": 47.266981,
      "label": "ci95",
      "low": 45.014132,
      "product": "gpt-6-luna-medium",
      "q": 46.140557,
      "rank": 34,
      "status": "active"
    },
    {
      "cost": 0.160127,
      "cost_confirmed": true,
      "high": 46.393353,
      "label": "ci95",
      "low": 43.64741,
      "product": "glm-5-3-flash",
      "q": 45.020381,
      "rank": 35,
      "status": "active"
    },
    {
      "cost": 0.631197,
      "cost_confirmed": true,
      "high": 47.734756,
      "label": "ci95",
      "low": 42.134841,
      "product": "step-5-preview",
      "q": 44.934799,
      "rank": 36,
      "status": "active"
    },
    {
      "cost": 0.241642,
      "cost_confirmed": true,
      "high": 46.693714,
      "label": "ci95",
      "low": 40.440989,
      "product": "minimax-m3-1-flash",
      "q": 43.567351,
      "rank": 37,
      "status": "active"
    },
    {
      "cost": 0.430487,
      "cost_confirmed": true,
      "high": 43.084554,
      "label": "ci95",
      "low": 39.7127,
      "product": "qwen3-8-27b",
      "q": 41.398627,
      "rank": 38,
      "status": "active"
    },
    {
      "cost": 0.221456,
      "cost_confirmed": true,
      "high": 42.779762,
      "label": "ci95",
      "low": 39.884921,
      "product": "qwen3-8-flash-next",
      "q": 41.332341,
      "rank": 39,
      "status": "active"
    },
    {
      "cost": 0.632693,
      "cost_confirmed": true,
      "high": 43.535596,
      "label": "ci95",
      "low": 38.006536,
      "product": "gemini-3-1-pro-preview",
      "q": 40.771066,
      "rank": 40,
      "status": "active"
    },
    {
      "cost": null,
      "cost_confirmed": false,
      "high": 45.736055,
      "label": "ci95",
      "low": 35.478759,
      "product": "beam-1-0-501b-a23b",
      "q": 40.607407,
      "rank": 41,
      "status": "active"
    },
    {
      "cost": 0.010452,
      "cost_confirmed": true,
      "high": 39.719359,
      "label": "ci95",
      "low": 36.79189,
      "product": "gpt-6-luna-low",
      "q": 38.255624,
      "rank": 42,
      "status": "active"
    },
    {
      "cost": 0.445551,
      "cost_confirmed": true,
      "high": 41.407183,
      "label": "ci95",
      "low": 32.954264,
      "product": "mistral-large-4",
      "q": 37.180723,
      "rank": 43,
      "status": "active"
    },
    {
      "cost": 0.353168,
      "cost_confirmed": true,
      "high": 39.058923,
      "label": "ci95",
      "low": 34.263497,
      "product": "qwen3-7-max",
      "q": 36.66121,
      "rank": 44,
      "status": "active"
    },
    {
      "cost": 0.14057,
      "cost_confirmed": true,
      "high": 38.499268,
      "label": "ci95",
      "low": 32.03128,
      "product": "qwen3-7-plus",
      "q": 35.265274,
      "rank": 45,
      "status": "active"
    },
    {
      "cost": 0.073335,
      "cost_confirmed": true,
      "high": 37.807241,
      "label": "ci95",
      "low": 32.054089,
      "product": "hy3",
      "q": 34.930665,
      "rank": 46,
      "status": "active"
    },
    {
      "cost": 0.372478,
      "cost_confirmed": true,
      "high": 35.966836,
      "label": "ci95",
      "low": 33.33192,
      "product": "muse-spark-1-1",
      "q": 34.649378,
      "rank": 47,
      "status": "active"
    },
    {
      "cost": 0.438531,
      "cost_confirmed": true,
      "high": 36.201892,
      "label": "ci95",
      "low": 32.832457,
      "product": "muse-spark-1-2-xhigh",
      "q": 34.517175,
      "rank": 48,
      "status": "active"
    },
    {
      "cost": 0.076846,
      "cost_confirmed": true,
      "high": 34.003631,
      "label": "ci95",
      "low": 29.949232,
      "product": "mimo-v2-5-pro",
      "q": 31.976431,
      "rank": 49,
      "status": "active"
    },
    {
      "cost": 0.097772,
      "cost_confirmed": true,
      "high": 32.97231,
      "label": "ci95",
      "low": 27.01375,
      "product": "spark-x2-6",
      "q": 29.99303,
      "rank": 50,
      "status": "active"
    },
    {
      "cost": 0.23518,
      "cost_confirmed": true,
      "high": 31.516786,
      "label": "ci95",
      "low": 26.82212,
      "product": "minimax-m3",
      "q": 29.169453,
      "rank": 51,
      "status": "active"
    },
    {
      "cost": 0.469507,
      "cost_confirmed": true,
      "high": 26.960364,
      "label": "ci95",
      "low": 20.174465,
      "product": "mistral-medium-3-5",
      "q": 23.567415,
      "rank": 52,
      "status": "active"
    },
    {
      "cost": 0.149715,
      "cost_confirmed": true,
      "high": 25.229886,
      "label": "ci95",
      "low": 12.363133,
      "product": "nemotron-3-ultra",
      "q": 18.796509,
      "rank": 53,
      "status": "active"
    },
    {
      "cost": null,
      "cost_confirmed": false,
      "high": 20.11268,
      "label": "ci95",
      "low": 16.226446,
      "product": "nvidia-nemotron-3-5-super-vl-120b-a12b-nvfp4",
      "q": 18.169563,
      "rank": 54,
      "status": "active"
    },
    {
      "cost": 0.332419,
      "cost_confirmed": true,
      "high": 20.801526,
      "label": "ci95",
      "low": 14.864282,
      "product": "inkling",
      "q": 17.832904,
      "rank": 55,
      "status": "active"
    },
    {
      "cost": 0.18831,
      "cost_confirmed": true,
      "high": 19.948368,
      "label": "ci95",
      "low": 12.485999,
      "product": "inkling-small",
      "q": 16.217183,
      "rank": 56,
      "status": "active"
    },
    {
      "cost": 0.05355,
      "cost_confirmed": true,
      "high": 12.308342,
      "label": "ci95",
      "low": 3.912338,
      "product": "solar-pro-4",
      "q": 8.11034,
      "rank": 57,
      "status": "active"
    }
  ],
  "schema": 1,
  "seed": null,
  "site": {
    "board_note": "Arena data: Arena (lmarena-ai), dataset lmarena-ai/leaderboard-dataset, https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset, licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Changes by Inadviser: the Arena scores (the Agent Arena IPS score and the text, text-to-image, text-to-video and search ratings) are rescaled to a 0-100 utility against fixed anchors set by Inadviser; the confidence interval and the observation or vote count are carried over; only a selection of models is shown on each board; where a board shows Arena's cost per task, it is shown unchanged. The dataset maintainers' method applies (https://arena.ai/blog/agent-arena-methodology/ for Agent Arena). Arena does not endorse Inadviser. Only one origin (Arena) is used, so no board entry can be named Best until a second independent origin or an own test exists. Arena's IPS score comes from agent sessions of its own users; it is not a test of this board's job on your code.",
    "first_paragraph": "As of 5 November 2026, no option is proven ahead. 3 options of 57 cannot be separated with the evidence in hand: Claude Opus 5.5 (Max), Gemini 4 Carbon (High) and Claude Sonnet 5.5 (Max).",
    "interval_labels": {
      "ci95": "built from 95% confidence intervals given by the source"
    },
    "ranking_is_quality_order": true,
    "ranking_note": "Order is by score point estimate; intervals may overlap.",
    "sources_used": [
      {
        "checked_on": "2026-11-03",
        "id": "arena-leaderboard-dataset",
        "name": "Arena Leaderboard Dataset (lmarena-ai/leaderboard-dataset)",
        "status": "attribution",
        "url": "https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset"
      }
    ],
    "verdict_label": "Shortlist"
  },
  "tariff_ids": [],
  "weights": {
    "agent_arena_ips_score": 1
  }
}
